Contemporary tests on software are performed in large labs that have multiple computers. To provide test data from a variety of platforms, testing may be performed on a variety of computers having different platforms (e.g., operating systems and upgrade levels), and/or languages. To limit the number of computers that are to be used in the lab, some of the individual computers may have several different platforms that are available on the computers, so that tests may be conducted on the same computers in multiple platforms.
To test the performance of a task within a particular software application, the software is loaded (if not already present) into a number of the computers and operators perform the task on each of the computers. For example, an application may be loaded and a particular user interface of the application may be displayed. Attempts may be made to display the user interface in several different platforms and/or languages. In the process of testing multiple platforms, a technician may run the task in one platform on a computer, reboot the computer to another platform, and run the test again. In addition, multiple computers having multiple platforms may be available, and technicians may perform the task on each of the multiple computers. In either situation, although a single technician may be able to perform tests on multiple machines, simultaneously performing tasks on multiple computers is an expensive process, requiring a large number of technicians to maintain several machines. In practice, only a subset of the available computers is used, and other computers sit idle between test processes, which wastes resources.